Background of the invention
1. Field of the invention
[0001] The invention pertains to a method for polarizing materials to produce electrets
for devices.
2. Art background
[0002] Electrets are used in a wide variety of devices such as transducer devices, transformers,
electric motors, and xerographic copying machines. (Electrets, for purposes of this
disclosure, are electrically polarized bodies whose polarization persists after being
produced). The electret-containing transducer devices include, for example, microphones,
loudspeakers, pressure sensors, touch sensitive keyboards and heat sensors. Furthermore,
transducer devices such as electrostatic microphones or loudspeakers often include
a thin film electret as the vibrating element of the device.
[0003] Materials employed in electrets are not inherently polarized but are polarizable.
There are a wide variety of techniques for producing the desired polarization. For
example, charges are injected into charge traps within or on the surface of the polarizable
material, free charges are separated within the material, or dipoles of the material
are aligned. Materials suitable for electrets and polarizable by charge injection
techniques include polyesters such as polyethylene terephthalate and fluorocarbons
such as fluorinated polyolefins. Polyvinylidene fluoride (PVDF) is exemplary of electret
materials which are polarized by aligning dipoles of the material. Polarized PVDF
is a commonly employed electret material because it exhibits both piezoelectric (an
electrical signal is produced in response to an applied stress) and pyroelectric (an
electrical signal is produced in response to heat) properties.
[0004] One technique for polarizing a body, e.g., a sheet, of polarizable material for electret
fabrication involves placing the body between two conducting, e.g., metal, electrodes,
and applying a voltage across the electrodes. If the body includes free charges or
unoriented dipoles, then the electric field existing between the electrodes induces
the charge separation or dipole alignment necessary to achieve polarization. If the
body includes charge traps, then placing the body in contact with one of the metal
electrodes results in a flow of charges from the electrode into the charge traps,
also producing polarization.
[0005] One problem often encountered with the previously described polarization technique
is large-scale dielectric breakdown. This typically results in the destruction of
a significant portion of the polarizable body (a hole several millimeters in diameter
is burned through the body) and results in either no polarization or an undesirably
low degree of polarization in the undamaged portions of the body. It is believed that
breakdown is generally initiated in a region of the body exhibiting a defect leading
to reduced electrical resistance, e.g., a pinhole, decreased thickness, or low dielectric
strength. A sufficiently large, applied polarizing voltage produces a short circuit
(of charges distributed on the surfaces of the electrodes facing the sheet) through
the defective region, reducing the polarizing voltage across the nondefective portions
of the sheet, and thus precluding effective polarization. The short circuit also involves
an avalanche-like effect which first produces a hole typically several micrometers
in diameter through the body, and ultimately leads to catastrophic failure.
[0006] A method for mitigating the effects of breakdown as described by T. T. Wang and J.
E. West in "Polarization of poly(vinylidene fluoride) by application of breakdown
fields", Journal of Applied Physics, 53 (10),1982, involves inserting a layer of dielectric
material, e.g., a layer of soda- lime glass, between the body of material to be polarized
and one of the conducting electrodes. (The dielectric insert either directly contacts
one surface of the sheet, or a metallized surface of the sheet). Breakdown, if it
occurs, is thus limited to a relatively small region, typically no more than a few
micrometers in diameter, and polarization occurs in the remaining portions of the
sheet.
[0007] It is established doctrine that the dielectric insert functions as distributed resistive
elements, and that only such distributed resistance prevents large-scale breakdown.
That is, each volume of the dielectric insert resistively limits current flow through
an adjacent coextensive volume of the polarizable material. Thus, the dielectric insert
allows a higher polarizing voltage across each volume without breakdown and results
in a concomitant enhanced degree of polarization. Additionally, the dielectric insert
resistively inhibits lateral current flow along the insert-electrode interface (provided
the sheet is unmetallized, or is metallized and has suffered a local breakdown producing
a discontinuity in the metal adjacentthe breakdown region). Thus, in the event of
a local breakdown, adverse consequences are limited by preventing charges distributed
on the electrode surface from flowing along the insert-electrode interface to the
local breakdown region.
[0008] While the use of a dielectric insert has many advantages, improvements are possible.
For example, a greater degree of polarization uniformity, and shorter polarization
times (the time required to achieve the desired degree of polarization) are desirable.
[0009] The preambles of claim 1 and 4 are based on EP-A-0014043 which discloses a process
for manufacturing polarized film in which a polarizable film is clamped between two
belts of conductive elastomer between which a voltage is applied. One of the belts
is continuous and is connected directly to one side of a power supply, whilst the
other belt is formed from electrically isolated segments. These segments are connected
via respective resistors to the other side of the power supply. By this arrangement,
if a defect in the film causes the belts to be short-circuited, only the region covered
by the segment involved fails to be polarized.
Summary of the invention
[0010] According to the invention there is provided a method as set out in claims 1 and
4.
[0011] We have found that a suitable reduction in the voltage reduces the likelihood of
breakdown and, surprisingly, still permits polarization to be achieved. Moreover,
polarization uniformity is enhanced and processing time is reduced, relative to the
dielectric insert technique.
[0012] One apparatus for implementing the inventive polarization method includes a conventional
feedback control device which automatically reduces the applied polarizing voltage
in response to the onset of a breakdown. An alternative device includes a discrete
(rather than distributed) resistor in the closed circuit path between a voltage source
and the material being polarized. If a breakdown is initiated in the polarizable material,
then there is a related increase in the current normally flowing through the resistor
during polarization. Consequently, the voltage across the material is necessarily,
and automatically, reduced by the increase in the voltage drop across the resistor.
Brief description of the drawings
[0013] Some embodiments of the invention are described with reference to the accompanying
drawings wherein:
Figs. 1-2 are schematic diagrams of apparatus for implementing the inventive fabrication
method;
Fig. 3 is a plot of the current, as a function of time, which typically flows through
a polarizable body during polarization, in the absence of breakdowns;
Fig. 4 depicts polarization data obtained using a dielectric insert; and
Fig. 5 is a schematic diagram of a polarization apparatus, which includes a discrete
resistor in place of a dielectric insert, used to obtain the polarization data shown
in Figs. 6-7.
Detailed description
[0014] The electret (of the electret-containing device) is formed by applying a voltage
across a polarizable body. This voltage is applied, for example, by placing the polarizable
body 10 (see Fig. 1) between electrically conductive, e.g., metal, electrodes 12 and
14, and applying a voltage across the electrodes. However, and unlike the dielectric
insert technique, the likelihood of breakdown in the polarizable body 10 is reduced
by reducing the applied voltage across at least a portion, and preferably all, of
the body 10 undergoing polarization in response to the onset of a breakdown in the
body 10. The size of the portion, e.g., the length or width of the portion, is preferably
larger than, or equal to, about 10 µm to avoid undesirably large nonuniformities in
the resulting polarization (achieved under the influence of the reduced voltage) of
the portion.
[0015] The onset of a breakdown is readily detected by sensing (using conventional techniques)
the current flowing through the (initially unpolarized or partially polarized) body
10 during polarization, and comparing this current with a reference current which
would flow in the absence of breakdown. A typical plot of a current exhibiting no
breakdown, as a function of time, is shown in Fig. 3, and is labelled i,. This current
consists of the sum of a current labelled i
2 in Fig. 3, associated with the capacitance, C, of the body 10, and a current labelled
i3 in Fig. 3, associated with the production of persistent polarization. To determine
when the voltage should be reduced, a plot of the current i
i which would flow in the absence of breakdown is obtained, prior to the polarization
of the (unpolarized or partially polarized) body 10, by subjecting a statistically
significant number of unpolarized (or partially polarized) control samples to the
polarizing voltage, and graphing the average of the resulting, measured currents as
a function of time. The current i
2 is obtained by subjecting these same, polarized (to the desired degree) control samples
to the same voltage, and again graphing the average of the resulting, measured currents.
The i3 current is just the difference of these two, average currents. It has been
found that the onset of a breakdown is signalled by a sharp, upward deviation in the
monitored current from the reference i, current. A deviation from the reference current
indicating the onset of breakdown is an amount greater than the maximum amplitude
of the i3 curve (denoted 1
3 in Fig. 3). Generally, such an amount is reached within a period of time less than
about 10 milliseconds (10x 10-3 seconds), and typically less than about 200 nanoseconds
(200xlO-
9 seconds), after the initiation of the deviation.
[0016] A voltage reduction is effective in reducing the likelihood of a breakdown if: (1)
it is large enough to reduce the deviation (from the reference current) to a size
less than, or equal to, the peak amplitude of the i3 curve; and (2) this reduction
in the amplitude of the deviation is effected within about 10 milliseconds, and preferably
within about 200 nanoseconds, and more preferably within about 20 nanoseconds, from
the time when the deviation from the reference curve exceeds the peak i3 amplitude.
Both the voltage reduction and the time period over which it is produced, needed to
achieve objectives (1) and (2), are generally determined empirically by monitoring
the onset of breakdowns in control samples and applying varying voltage reductions
over varying periods of time.
[0017] An apparatus for automatically reducing the applied voltage in response to the onset
of a breakdown includes, for example, a conventional feedback control circuit 16 (pictured
as a box in Fig. 1) which automatically reduces the voltage of a variable voltage
source 18 in response to a current deviation in the body 10, described above. While
the feedback control circuit 16 and the variable voltage source 18 are pictured as
being separate components, in practice they are often manufactured as a single, integral
unit.
[0018] A particular type of control circuit which automatically reduces the applied voltage
in response to the onset of a breakdown is schematically depicted in Fig. 2 and comprises
a resistor 20 (having an essentially zero response time) in the circuit path between
a constant voltage source 22 and the polarizable body 10. Because the body 10, electrodes
12 and 14, resistor 20 and voltage source 22 form a closed circuit loop, whatever
current flows through the body 10 must necessarily also flow through the resistor
20. Consequently, the increment in current flowing through the resistor 20 in response
to the onset of a breakdown in the body 10 necessarily results in an essentially instantaneous
voltage drop across the body equal to the increase in the voltage drop across the
resistor.
[0019] It has been found that the use of a resistor 20 is particularly advantageous because
a lower value of the resistance of the resistor reduces the polarization time. Consequently,
by choosing an appropriate resistor, the time required to achieve a desired degree
of polarization in a particular material, and for a particular applied voltage, is
generally less than when using a dielectric insert.
[0020] The resistance of the resistor20 is limited by two considerations. The first of these
is that the resistance should be large enough to prevent, or at least reduce the possibility
of, large-scale breakdown. The second consideration is that any voltage drop across
the resistor (if it occurs) should not be so large that the voltage across the polarizable
body 10 is undesirably reduced (the smaller the voltage across the body 10, the smaller
the degree of polarization). Within the bounds defined by these two considerations,
the resistance is chosen to take on any value. Typically, however, the resistance
is chosen, empirically (by applying polarizing voltages to control samples through
resistors of different resistance and measuring the resulting polarizations over varying
periods of time) to achieve the desired polarization within as small a time as is
compatible with other processing considerations.
[0021] The resistance of the resistor 20 is chosen in relation to the material properties
and dimensions of the body 10. In practice, the polarizable bodies 10 include materials
having resistivities greater than about 1x10
8 ohm-m and relative dielectric constants greater than about 1.5. Moreover, these bodies
are typically sheets having thicknesses ranging from about 1 µm to about 1 cm, and
surface areas (the multiple of the length and width) ranging from about 1 x 10
-10 m
2 to about 1 m
2. Consequently, these bodies have resistances greater than about 100 ohms and capacitances,
C, reater than about 1.5x10
-19 farads. Materials having resistivities less than about 1 x108 ohm-m or relative dielectric
constants less than about 1.5, are undesirable because they are difficult to polarize.
Sheet thicknesses less than about 1 µm are undesirable because the probability of
breakdown is undesirably high, while sheet thicknesses greater than about 1 cm are
undesirable because undesirably large voltages are needed to polarize such sheets.
Sheets having surface areas less than about 1 x10
-10 m
2 are undesirable because they are difficult to handle and exhibit undesirably large
polarization nonuniformities. Sheets having surface areas greater than about 1 m
2 are undesirable because they exhibit an undesirably large number of manufacturing
defects, e.g., pinholes.
[0022] For polarizable bodies having the above properties and dimensions, the resistance,
R, of the resistor 20 ranges from about 10 ohms to about 1 x 10
20 ohms, and preferably ranges from about 1 x 10
4 to about 1 x104 ohms. Resistances less than about 10 ohms are undesirable because
the likelihood of breakdown is undesirably high. Resistances greater than about 10
2° ohms are undesirable because they result in undesirably long polarization times
and result in undesirably small voltages across the polarizable body 10 in the event
of a local breakdown.
[0023] In the case of, for example, a sheet of PVDF (PVDF has a resistivity of about 1 x
1 012 ohm-m and a relative dielectric constant of about 12), the thickness ranges
from about 1 pm to about 1 cm, and the surface area ranges from about 1 x10
-10 m
2 to about 1 m
2, leading to resistances (of the sheet), which range from about 1 x 10
6 to about 1 x 10
20 ohms and capacitances which range from about 1x10
-18 to about 1x10
-4 farads. Thicknesses and surface areas outside the above ranges are undesirable for
the reasons given above. For a PVDF sheet having the above dimensions, the resistance
of the resistor 20 ranges from about 1x10
6 about 1x10
20 ohms. Resistances outside the above range are undesirable for the reasons given above.
[0024] While the above discussion has been limited to using a voltage reduction means, the
use of a voltage reduction in combination with other expedients such as a dielectric
insert is not precluded and is often desirable because it decreases the level of voltage
reduction required.
[0025] After the polarization of the body 10 to produce the electret, the device, which
includes the electret, is completed by a series of conventional steps (which differ
depending on the nature of the device), as described, for example, by G. M. Sessler
and J. E. West in "Electrets", Topics in Applied Physics, 33, G. M. Sessler, ed.,
(Springer, New York), 1980.
Example 1
[0026] For purposes of comparison, polarization results obtained with a dielectric insert
are described below.
[0027] Square PVDF sheets, each 9 cm in width and 9 cm in length, were cut from a roll of
25 pm-thick, biaxially oriented (the material had been stretched in two perpendicular
directions) capacitor-grade film produced by the Kureha Chemical Company of Tokyo,
Japan. These sheets were cleaned with trichloroethane, and then circular aluminum
electrodes, about 100 nm in thickness and 6.5 cm in diameter, were vacuum deposited
onto opposite faces of each sheet, leaving an unmetallized border to prevent flashover
(arcing) during the ensuing polarization procedure.
[0028] The PVDF sheets were placed (one at a time) on a square plate of soda lime glass
(the dielectric insert) having a thickness of 0.24 cm and a length and width of 15
cm, arranged between two circular brass electrodes. Each electrode had a thickness
of 1.3 cm and a diameter of 6.5 cm. The PVDF sheets were then polarized (at room temperature)
by applying a dc voltage of 10, 15, 18.5, 21.5, 24.5 or 30 kilovolts across the brass
electrodes, for periods of time ranging from about 2.5 minutes to about 3 hours.
[0029] After being polarized, the PVDF sheets were stored, at room temperature, in a short-circuit
condition (the circular aluminum electrodes were electrically connected to one another)
for about 16 hours.
[0030] Upon removal from storage, the degree of polarization achieved by each PVDF sheet
was determined by measuring the piezoelectric activity, i.e., the piezoelectric strain
coefficient, d
31, of each sheet. (d
31=∂ P
3/∂σ
11 where P
3 denotes the polarization in the "3" direction and σ
11denotes the stress applied in the "1" direction to a surface whose normal is parallel
to the "1" direction. As is conventional, the "3" direction is parallel to the surface
normal of each sheet while the "1" direction is aligned with the longitudinal direction
of the as-received roll of film). The piezoelectric measurements were performed by
applying a sinusoidal strain with a frequency of 110 Hz and an amplitude of 0.036
percent (regarding this technique see, e.g., T. T. Wang, Journal of Applied Physics,
50, 6091 (1979)). The measured results are depicted in Fig. 4 which is a plot of the
apparent piezoelectric stress coefficient, e
31 (e
31 =E
11 d
31, where E
11 is the Young's modulus for PVDF) as a function of polarization time, tp, for the
various voltages employed. For the maximum applied voltage of 30 KV, the e
31 coefficient reaches a value of 29 mC/m
2 after 30 minutes.
Example 2
[0031] The polarization results obtained using an external resistor, rather than a dielectric
insert, are described below.
[0032] Square PVDF sheets, each 12 cm in width and 12 cm in length, were cut from a roll
of 25 µm-thick, biaxially oriented capacitor-grade film produced by the Kureha Chemical
Company. These sheets were cleaned with trichloroethane, and then circular aluminum
electrodes, about 100 nm in thickness and 6 cm in diameter, were evaporated onto the
opposite faces of each sheet, leaving an unmetallized border area at least 3 cm wide.
[0033] The apparatus used to polarize each PVDF sheet is depicted in Fig. 5 and was similar
to that used in Example 1 except for the absence of the plate of soda lime glass (the
dielectric insert) and the insertion of one of several resistors, purchased from the
Beman Company of New Cumberland, Pennsylvania, between the power supply and one of
the brass electrodes. The apparatus also included a lower plate of polymethyl methacrylate
(PMMA) supporting the lower brass electrode, as well as an upper PMMA plate resting
on a silicone O-ring supported by the PVDF sheet, which included a central aperture
allowing the plate to clear the upper brass electrode. The purpose of the PMMA plates
was to prevent flashover (arcing) around the sheet edges.
[0034] The PVDF sheets were polarized (at room temperature) by applying a dc voltage ranging
from 5 to 30 KV across the brass electrodes, for periods of time ranging from 1 to
1 x 10
5 seconds, using resistors whose resistances, R, ranged from 5x10
7 ohms to 1 x 10
11 ohms.
[0035] After being polarized, the PVDF sheets were stored in a short-circuit condition at
room temperature for more than 72 hours, and cut into rectangular pieces having lengths
of 0.45 cm and widths of 5.3 cm. The piezoelectric activity of each sheet was then
measured, as in Example 1.
[0036] A plot of the apparent piezoelectric stress coefficient, e
31, as a function of polarization time, tp, obtained at various applied voltages, with
R=
1x10
9 ohms, is depicted in Fig. 6. It can be seen that while the e
31 for each applied voltage rises with polarization time, it tends to saturate at a
progressively higher value with increasing voltage. The maximum e
31 value obtained at 30 KV is 29 mC/m
2, which is about the same maximum obtained in Example 1 using the same voltage. However,
a polarization time of only 25 seconds was needed to achieve the maximum e
31 in the present case, whereas 30 minutes was needed to achieve this maximum in Example
1.
[0037] Comparison of other e
31 data in Fig. 6 with the corresponding results obtained in Example 1 shows that the
time required to achieve the same e
31 value at a given applied voltage is always shorter using the present method than
the method used in Example 1.
[0038] The e
31 data obtained from samples polarized with an applied voltage of 5 KV, but using resistors
with R ranging from 5x10
7 to 1x10
11 ohms Fig. 7. Also shown are the corresponding data for PVDF sheets prepared using
R<1x10
5 ohms (regarding this data see T. T. Wang and H. von Seggern, Journal of Applied Physics,
54, 4602 (1983)). As is evident, the e
31 curve for each R rises with polarization time and gradually approaches the corresponding
curve for R<1 x 10
5 ohms. This is due to the fact that the same voltage was applied in each case, and
thus the same degree of polarization, and therefore the same e
31 coefficient, is. ultimately (after a long enough polarization time) achieved in each
case. Moreover, all the e
31 curves seem to have about the same shape except for a horizontal shift which increases
as R increases. That is, as R increases, the time required to achieve the same degree
of polarization, i.e., same value of e
31, increases.